Floyd Sweet VTA Replication Project

[mx1000]

Ok, I had to give it a try. I had one crazy thought, or was it an inspiration by sparky himself , appearing in a dream...


2 microwave oven magnets, two coils. The idea was to make the inner coil in a special way, so it would be capable of guiding the flux in a passive state...


So far so good, I tried it with a 12 V AC wave, both, full and a halfrectified "dc pulse with 50% duty time". Output was 1.4v with no measurable amps, practically zero and the led load didn't light up at all. Even with the full power of the supply, normally 20 Watts absolutely no output current.


Then I tried it with a squareish signal of 12.5% duty time, using the pc and a little LM386 circuit as an amplifer. Unstantly the LED lit up. I increased the frequency and found a max amplitude at about 1250 hz.


I triwd to measure in and out. Volts in parallel and Amps in series with the LED, with a electrolytic cap, 1000uF 16V, in parallel with the LED:
2.7V * 2.6mA = 7.02 mW


I then connected the LM386 output directly to the VTA output  practically skipping the VTA. I read 1.98Volts and really 0.000 mA. This is bad for calculation, so I took a 10 ohm resistor instead of the LED, but still with the cap:
0.16V * 9mA = 1.44 mW


This sounds too good, and indeed, measuring the power when the 10 ohm resistor was connected to the VTA gave me dissapointing 0.09V @ 8mA,


So you see, here seems to be sorta discrepancy, depending on the load. Not shure what I should think about, but I thought I let you know.


Remember, Sparkys first model that he presented to Tom Bearden had an Output of 6 Watt..
Some pics...

Greatly appreciated.

I can't add much, although I have a link of a lot of Floyd Sweet vids and stuff (a Russian site).
If you want I can post them here.

Thank you for sharing.

[MenofFather]

Great work.

[dieter]

Thank you very much, your feedback is highly appreciated.

mx1000, shure, post the links.

As an update: by modelling some waveforms with rather steep increase and then sudden fall to zero within 1 sample of the pcm wav (using a 22050 hz wav, but can play it at any rate),  I increased the amps to 3.4 mA, so now that is 9mW output. The described problems in measuring the input remain a problem. All I know is that I get these 9mW as a serious, smoothed current from the cap, pure dc.

Now, as I was not able to light the blue led by connecting the rectifier and cap directly to the LM386, i tried a tiny red LED, made for 20mA@2v (the blue one has 5 parallel LEDs integrated , parallel, for 3 volts). So this tiny red LED hooked up to the LM386 speaker output did light up with a nice brightness. I tried to add a cap, but that just turned it off, due to the AC nature of the signal. I measured: 1.19V, 0.04 mA (on 2.000 mA meter range) . I multiplied this by 4 since in AC the voltage is 2x and the current flows in 2 directions (maybe x2 would have been enough, anyway). So, as a very rough estimation I had 0.188 mW here. What can  I say, I  could hardly believe it. Also remarkable was, I was getting a moderate Voltage gain under load of maybe 200%, but the current gain was the amazing thing. 0.188mW in, 9mW out.

It certainly had something to do with the back EMF since the waveform was optimized for it.

But let me go a little deeper into the construction details. The basic idea for the concept, that made me try this in the first place, is something simple yet mind bending: The inner coil, that has a 5mm aircore and is of high mass, is no copper coil at all, it is a IRON coil. It is fence wire, covered with green color, so it is insulated.

The idea was that when the signal is idle between pulses, the magnets would use this mass of iron like a core, flow trough it in a low leakage flux path, reducing the field strength around it. Then, durind the pulse that flows trough the iron coil, that decreases the permeability of the iron mass due to an opposite field, the permanent magnets would avoid the iron as a flux path, preferring to flow trough the space around it, basicly trough the outer copper coil.

I don't know if that really happens, but I made a very fascinating observation with this iron coil: It is attracted from a magnets north side much more than from the south side !!!  It's just a piece of iron, but yet it does so! Why is that so? Well, since it is a coil, a current will be induced when you move it to a magnet. That current causes a magnetical field, that will add or subtract to the natural attraction of iron, depending on the perm. magnets polarity. So this way we have just created partially monopolar ferromagnetism. Not shure about what it's good for, but it is fascinating and I will investigate this further. Maybe it has something to do with the strange output gain. I may however try to use a real amplifer and pump like 100 Watts or so into it, instead of the LM386 that dissipates only 1 watt at max.

Regards

[dieter]

Just for the records,

...
Why is that so? Well, since it is a coil, a current will be induced when you move it to a magnet. That current causes a magnetical field, that will add or subtract to the natural attraction of iron, depending on the perm. magnets polarity. So this way we have just created partially monopolar ferromagnetism...

Of course, this was nonsense. No matter which winding direction, the lorentz force will always act against the  mechanical force that causes the induction, that's Lenz law 101. I was pretty much out of my head that day. So please excuse.

It turned out, the magnets are magnetized in a special way, in that any  piece of  iron sticks more on one side than on the other.

Regards